For highlighting a shot subject and focusing on the shot subject in a shot image, real-time blurring may usually be performed on a background region in the shot image.
At present, when real-time blurring is performed on a background region of a shot image, depth information of field of a frame picture is usually calculated in a process of displaying the frame picture acquired by an image sensor. Then blurring is performed on a part of the background region according to the calculated depth information of field. However, when blurring is performed on the shot picture in such a manner, relatively long time is required for calculation of the depth information of field of the shot picture. Moreover, when a shot subject shakes, there is no enough time to update the depth information of field of the shot picture so as to make it likely to blur part of the subject, which further brings influence on an overall blurring effect and worsens user experience.